Vision API should allow detecting multiple "feature types" in a single request

[tswast]

See: https://cloud.google.com/vision/pricing

Running "Explicit Content Detection" is free if done in the same request as "Label Detection". It appears from https://googlecloudplatform.github.io/google-cloud-python/stable/vision-usage.html that running image.detect_safe_search() immediately runs the request without having the chance to say I also want to run image. detect_labels() in the same HTTP request (to save money and an HTTP round-trip).

[daspecster]

Thanks for letting me know!

@tswast, we generally call this "batching" in the library. It should be in shortly after a refactor for adding manual detect().

This can be tracked in #2753.

I actually should have the PR for detect() tonight and then I think I have a good plan for batching as well.

[daspecster]

I guess I used the wrong syntax for Closes in the PR.

https://help.github.com/articles/closing-issues-via-commit-messages/#closing-multiple-issues

[ilsePit]

Is this also possible in the new version (v0.25.1 and up)? I cannot find this functionality in the documentation.